CN114888816A - Control system and method of intelligent feeding and discharging robot - Google Patents

Abstract

The invention provides a control system and a method of an intelligent loading and unloading robot, wherein the system comprises: the acquisition module is used for acquiring the feeding area; and the control module is used for controlling the robot to transfer the workpieces in the feeding area to the input end of the processing area, the feeding is completed, and after the processing is completed, the robot is controlled to transfer the workpieces at the output end of the processing area to the preset discharging area. The control system and the method of the intelligent loading and unloading robot have the advantages that workers do not need to be arranged beside the input end and the output end of the processing equipment, labor cost is reduced, in addition, when some workpieces need to be in butt joint with the fixing device of the input end of the processing equipment during loading, the workpieces can be loaded by the robot, and the completion efficiency is improved.

Description

Control system and method of intelligent feeding and discharging robot

Technical Field

The invention relates to the technical field of robots, in particular to a control system and a control method of an intelligent loading and unloading robot.

Background

At present, when some processing equipment (such as a cleaning machine) needs to carry out loading and unloading (such as inputting a workpiece needing to be cleaned into the cleaning machine, and taking out the workpiece after cleaning is finished), workers are mostly arranged beside an input end and an output end of the processing equipment and are manually finished by the workers, so that the labor cost is high. In addition, when some workpieces are loaded, the workpieces need to be butted with a fixing device (such as a fixing seat) at the input end of the processing equipment, and the manual finishing efficiency is low.

Therefore, a solution is needed.

Disclosure of Invention

The invention provides a control system and a control method of an intelligent loading and unloading robot, which do not need to arrange workers beside an input end and an output end of a processing device, reduce the labor cost, and can be completed by the robot when some workpieces need to be butted with a fixing device at the input end of the processing device during loading, thereby improving the completion efficiency.

The invention provides a control system of an intelligent loading and unloading robot, which comprises:

the acquisition module is used for acquiring the feeding area;

the control module is used for controlling the robot to transfer the workpieces in the feeding area to the input end of the processing area, the feeding is finished, and after the processing is finished, the robot is controlled to transfer the workpieces at the output end of the processing area to a preset discharging area;

the acquisition module acquires a loading area, including:

continuously acquiring a first image in a preset range around the robot;

determining a first position block of a residual historical loading area in a preset range around the robot based on the first image;

determining the distribution of the feeding areas of the remaining historical feeding areas within a preset range around the robot based on the first position block;

acquiring the feeding state of a historical feeding area;

classifying the historical loading area into a loaded target and an unloaded target based on the loading state;

planning at least one first route of the loaded target to leave a preset range based on the first position block of the loaded target, the distribution of the loading area and a preset range map;

manufacturing an evasion route set based on the first route;

selecting a third position block which is idle and has no way for all second routes in the avoidance route set and has a first straight line distance between a first center position of the second position block and the robot smaller than or equal to a preset straight line distance from a plurality of second position blocks in the range map;

selecting a third position block corresponding to the minimum second straight-line distance in the second straight-line distances between the second center position of the first position block without the fed target and the first center position of the third position block as a target position block;

determining at least one loading person who enters a preset range around the robot and uses the loading carrier based on the current first image;

reminding a loading person to move the loading carrier to any target position block;

and determining that the loading personnel moves the loading carrier to the target position block based on a first image acquired after reminding the loading personnel to move the loading carrier to any target position block, and finishing acquisition as a loading area.

Preferably, the obtaining module makes the avoidance route set based on the first route, and includes:

acquiring the overlapping degree between a first route which is left by any one loaded target and a first route which is left by other loaded targets, and associating the overlapping degree with the corresponding first route which is left by the loaded target;

accumulating and calculating the overlapping degree associated with the first route from which the fed target leaves to obtain the overlapping degree sum of the first route from which the fed target leaves;

taking the overlapping degree and the maximum overlapping degree of the first route from which each fed target leaves and the corresponding first route as a third route;

and manufacturing an avoidance route set based on each third route.

Preferably, the acquisition module reminds the loading personnel to move the loading carrier to any target position block, including:

determining the face position and the face orientation of the loading person based on the current first image;

constructing a first direction vector based on the face position and the face orientation;

acquiring equipment positions and equipment orientations of a plurality of first prompt equipment in a preset range around the robot;

constructing a second direction vector based on the device location and the device orientation;

calculating a first included angle between the first direction vector and the second direction vector;

selecting a first prompting device corresponding to the maximum first included angle as a second prompting device;

determining a third direction vector based on the device location of the second prompting device and the third center location of the target location block;

on the basis of a preset direction vector-guide information comparison library, comparing and determining first guide information corresponding to a third direction vector;

and controlling the second prompting device to display the first guide information.

Preferably, the acquisition module reminds the material loading personnel to move the material loading carrier to any target location piece, include:

acquiring projection ranges of a plurality of projection devices in a preset range around the robot;

when the projection range comprises the position area range of the target position block, acquiring a projection route corresponding to the projection equipment when the projection equipment projects preset second guide information in the position area range of the target position block;

determining whether an occlusion exists on the projection route based on the current first image;

if not, controlling any corresponding projection equipment to project second guide information in the position area range of the target position block.

Preferably, when the control module controls the robot to transfer the workpiece in the loading area to the input end of the processing area or transfer the workpiece at the output end of the processing area to the preset unloading area, when the manipulator of the robot is ready to clamp the workpiece, the best grabbing force of the workpiece is obtained, and the robot is controlled to grab the workpiece with the best grabbing force.

Preferably, the control module obtains the best grabbing strength of the workpiece, and the method comprises the following steps:

acquiring a workpiece image of a workpiece;

determining structural information and materials of the workpiece based on the workpiece image;

extracting a plurality of structural features of the structural information based on a preset first feature extraction template;

training an optimal grabbing force determination model;

and inputting the material and the structural characteristics to the optimal grabbing force determination model based on the optimal grabbing force determination model, and determining the optimal grabbing force.

Preferably, the control module trains an optimal grasping force determination model, including:

a plurality of grasping strength test records obtained from a big data platform and/or from the local;

preprocessing a gripping force test record;

taking the preprocessing result as a training sample, and carrying out model training on a preset neural network model;

after training is finished, taking the neural network model as an optimal grabbing force determination model;

wherein, control module carries out the preliminary treatment to snatching dynamics test record, includes:

acquiring record information of a grabbing force test record, wherein the record information comprises: source information and test procedure information;

extracting a plurality of recording features of the recording information based on a preset second feature extraction template;

acquiring a preset trigger feature library, and matching the recording features with first trigger features in the trigger feature library;

if the matching is in accordance with the first trigger characteristic, acquiring a trigger type of the first trigger characteristic in accordance with the matching, wherein the trigger type comprises: individual triggering and coordinated triggering;

when the trigger type is single trigger, acquiring a preset first verification scoring template corresponding to the matched first trigger characteristic;

when the trigger type is cooperative trigger, acquiring a preset cooperative trigger second trigger characteristic corresponding to the matched first trigger characteristic;

matching the recording features except the recording features matched and matched with the second trigger features in the recording features;

if the matching is in accordance with the preset verification marking template, acquiring a preset second verification marking template corresponding to the first triggering characteristic and the second triggering characteristic which are in accordance with the matching;

verifying and scoring the recording characteristics based on the first verification scoring template and the second verification scoring template, and accumulating scoring results to obtain a scoring sum;

if the score sum is less than or equal to the preset score sum threshold value, rejecting the corresponding grabbing force test record;

and after all the grabbing force test records needing to be removed are removed, integrating the remaining grabbing force test records to obtain a preprocessing result.

The invention provides a control method of an intelligent loading and unloading robot, which comprises the following steps:

step 1: obtaining a feeding area;

step 2: controlling the robot to transfer the workpieces in the feeding area to the input end of the processing area, completing feeding, and after the processing is completed, controlling the robot to transfer the workpieces at the output end of the processing area to a preset discharging area;

step 1: obtain the material loading district, include:

continuously acquiring a first image in a preset range around the robot;

determining a first position block of a residual historical loading area in a preset range around the robot based on the first image;

determining the distribution of the feeding areas of the remaining historical feeding areas within a preset range around the robot based on the first position block;

acquiring the feeding state of a historical feeding area;

classifying the historical loading area into a loaded target and an unloaded target based on the loading state;

planning at least one first route of the loaded target to leave a preset range based on the first position block of the loaded target, the distribution of the loading area and a preset range map;

manufacturing an evasion route set based on the first route;

selecting a third position block which is idle and has no way for all second routes in the avoidance route set and has a first straight line distance between a first center position of the second position block and the robot smaller than or equal to a preset straight line distance from a plurality of second position blocks in the range map;

selecting a third position block corresponding to the minimum second straight-line distance in the second straight-line distances between the second center position of the first position block without the fed target and the first center position of the third position block as a target position block;

determining at least one loading person who enters a preset range around the robot and uses the loading carrier based on the current first image;

reminding a loading person to move the loading carrier to any target position block;

and determining that the loading personnel moves the loading carrier to the target position block based on a first image acquired after reminding the loading personnel to move the loading carrier to any target position block, and finishing acquisition as a loading area.

Preferably, the making of the avoidance route set based on the first route includes:

acquiring the overlapping degree between a first route which is left by any one loaded target and a first route which is left by other loaded targets, and associating the overlapping degree with the corresponding first route which is left by the loaded target;

accumulating and calculating the overlapping degree associated with the first route from which the fed target leaves to obtain the overlapping degree sum of the first route from which the fed target leaves;

taking the overlapping degree and the maximum overlapping degree of the first route from which each fed target leaves and the corresponding first route as a third route;

and manufacturing an avoidance route set based on each third route.

Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.

The technical solution of the present invention is further described in detail by the accompanying drawings and embodiments.

Drawings

The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings:

FIG. 1 is a schematic view of a part of the process of the feeding zone acquisition.

Detailed Description

The preferred embodiments of the present invention will be described in conjunction with the accompanying drawings, and it will be understood that they are described herein for the purpose of illustration and explanation and not limitation.

The invention provides a control system of an intelligent loading and unloading robot, which comprises:

the acquisition module is used for acquiring the feeding area;

the control module is used for controlling the robot to transfer the workpieces in the feeding area to the input end of the processing area, the feeding is finished, and after the processing is finished, the robot is controlled to transfer the workpieces at the output end of the processing area to a preset discharging area;

the acquisition module acquires a loading area, including:

continuously acquiring a first image in a preset range around the robot;

determining a first position block of a residual historical loading area in a preset range around the robot based on the first image;

determining the distribution of the feeding areas of the remaining historical feeding areas within a preset range around the robot based on the first position block;

acquiring the feeding state of a historical feeding area;

classifying the historical loading area into a loaded target and an unloaded target based on the loading state;

planning at least one first route of the loaded target to leave a preset range based on the first position block of the loaded target, the distribution of the loading area and a preset range map;

manufacturing an evasion route set based on the first route;

selecting a third position block which is idle and has no way for all second routes in the avoidance route set and has a first straight line distance between a first center position of the second position block and the robot smaller than or equal to a preset straight line distance from a plurality of second position blocks in the range map;

selecting a third position block corresponding to the minimum second straight-line distance in the second straight-line distances between the second center position of the first position block without the fed target and the first center position of the third position block as a target position block;

determining at least one loading person who enters a preset range around the robot and uses the loading carrier based on the current first image;

reminding a loading person to move the loading carrier to any target position block;

and determining that the loading personnel moves the loading carrier to the target position block based on a first image acquired after reminding the loading personnel to move the loading carrier to any target position block, and finishing acquisition as a loading area.

The working principle and the beneficial effects of the technical scheme are as follows:

the feeding area specifically comprises: in the area where the workpieces needing to be loaded are placed in the field, the loading area can be input and set by robot debugging personnel. The treatment area is specifically as follows: in a processing apparatus for processing a workpiece (e.g., a cleaning machine for cleaning a workpiece) in a field, a processing region has an input end and an output end, and the input end is specifically: the workpiece enters the inlet of the processing equipment, and the output end is specifically as follows: and the workpiece is output after being processed by the processing equipment. The preset blanking area specifically comprises the following steps: the area in which the processed workpiece is placed in the field can also be input and set by a robot debugging person.

The manipulator of this application control robot shifts the work piece in the material loading district to the input of treatment area, and the work piece gets into the treatment area by the input, carries out corresponding processing by the treatment area, and after the processing was accomplished, the work piece was come out by the output of treatment area, and the manipulator of control robot shifts the work piece of the output of treatment area to in the unloading district of predetermineeing. In addition, when the workpiece needs to be fixed on a fixing device (for example, a fixing seat on a cleaning conveying belt of a cleaning machine) at the input end of the processing area, corresponding debugging setting can be carried out by a robot debugging person.

In addition, when some workpieces need to be butted with a fixing device at the input end of the processing equipment during feeding, the workpieces can be finished by a robot, and the finishing efficiency is improved;

the preset range is specifically as follows: the position of the robot is used as the center of a circle in the site, and the radius is 6 meters. The remaining historical loading areas are specifically: historically, when a worker transports a workpiece to a robot by using a loading carrier, the worker leaves the robot first to wait for the robot to carry objects in the loading carrier, and the rest historical loading area is the area where the loading carrier which is not pushed away is located. The first position block specifically comprises: the range of regional locations of the remaining historical loading zone. As shown in fig. 1, a in the figure is a position area of the robot, B is a preset range, and C1 and C2 are first position blocks of the remaining historical loading area. The feeding area is specifically distributed as follows: the location block distribution consisting of the first location blocks of each remaining historical feeding area, i.e. where there is a first location block, as shown in fig. 1, C1 and C2 constitute the feeding area distribution. The preset range map specifically comprises: the three-dimensional map corresponds to a circular range with the radius of 6 meters and the robot as the circle center, the arrangement positions of all the machine equipment in the site and the three-dimensional models of the equipment are arranged on the three-dimensional map, and the three-dimensional map can be drawn by workers in advance. The preset linear distance is specifically as follows: indicating a close distance to the robot, which may take 2.5 meters. The second position block is specifically: the staff member divides the range map into a plurality of location areas in advance according to the arrangement of the devices in the field, and D is a second location block as shown in fig. 1. The first center position is specifically: the position of the center point of the second position block. The second center position is specifically: the position of the center point of the first position block of the unloaded target.

Generally, the loading area is all fixed and set up, needs staff etc. to shift the work piece to the loading area, for example: the worker uses the loading carrier (e.g. cart) to transport the workpiece to the loading area, and directly transport the workpiece to the loading area, for example: the staff utilizes the material loading carrier to transport the work piece to the assembly line by side, on carrying the work piece to the assembly line again, transports the work piece to the material loading district by the assembly line, and very loaded down with trivial details, the human cost is great, and intelligent degree is lower. Especially when the number of workpieces to be processed is large.

In this application, the robot is fixed to be set up, and the material loading personnel only need utilize the material loading carrier to transport the work piece to the target position piece that is located the robot by side, and the people can leave, and the manipulator of robot shifts the input of treatment area to the work piece in with the material loading carrier, when waiting for the material loading personnel to utilize material loading carrier transportation other work pieces to come again, takes away the inside work piece of last use to the empty material loading carrier that has shifted the input of treatment area by the robot, and from this circulation. The convenience is greatly improved, and the labor cost is reduced.

In addition, when a loading person uses the loading carrier again to transport other workpieces, and wants to take away the workpiece inside the last time of use from the empty loading carrier transferred to the input end of the processing area by the robot, the empty loading carrier may not be taken away by the loading person due to limited space around the robot or unreasonable arrangement of a plurality of loading carriers, which may cause confusion and affect loading order and loading efficiency.

Therefore, the worker needs to push away the loaded target, and wants to ensure that the loaded target leaves the route when the worker who pushes the loading carrier carrying the new workpiece to come to the side of the machine recommends the loading area. Therefore, when the recommended feeding area is selected from the second position block, a first route of the fed target leaving the preset range is planned, and when the recommended feeding area is selected, all second routes in the evaded route set formed by the first routes are ensured to be not accessed, so that the fed target leaves the route. In addition, when selecting, guarantee that the first straight line distance between the first central point of second position piece and the robot is less than or equal to predetermined straight line distance for the material loading district recommended is nearer apart from the robot, promotes material loading efficiency. Secondly, during selection, the selected second position block is ensured to be idle, namely, no object is placed in the second position block, and the second position block can be determined according to the first image based on the image recognition technology. And selecting a second position block satisfying the three points as a third position block. In addition, the recommended loading area needs to be as close as possible to the unloaded target, and the moving stroke of the robot for switching to the next unloaded carrier for carrying and loading is reduced, so that the third position block corresponding to the minimum second straight-line distance in the second straight-line distance between the second center position of the first position block of the unloaded target and the first center position of the third position block is selected as the target position block, the target position block is recommended for a worker pushing the loaded carrier with a new workpiece to come beside the machine, and finally, the worker is determined to move the loaded carrier to the target position block and serve as the loading area based on the subsequent first image.

Assuming that C1 is the loaded target, the worker needs to push away to plan the first route E away from the preset range B, as shown in fig. 1. Assuming that C2 is an unloaded target, when selecting the third position block from the second position blocks D, the first route is ensured to be passing, and the third position block is close to the robot a and free and close to other unloaded targets C2, so that the second position block F is selected as the target position block, and the worker is recommended to push the loading carrier with the new workpiece to the target position block.

This application rationally for pushing away the material loading carrier that carries new work piece and coming to the other staff's of machine recommendation material loading district, avoid the material loading personnel can't take away empty material loading carrier, guaranteed material loading order and material loading efficiency, also promoted the rationality that the peripheral material loading carrier of robot was put, more saved the consumption of robot.

Secondly, when planning at least one first route of the loaded target leaving the preset range based on the first position block of the loaded target, the distribution of the loading area and the preset range map, taking the center of the first position block of the loaded target as a route planning starting point, and during planning, ensuring that the avoided route avoids other first position blocks except the loaded target in the distribution of the loading area, avoiding the shielding of other historical loading areas when the loaded target leaves, and also avoiding some field devices and the like in the field based on the range map.

The invention provides a control system of an intelligent loading and unloading robot.A obtaining module makes an evaded route set based on a first route, and the control system comprises:

acquiring the overlapping degree between a first route which is left by any one loaded target and a first route which is left by other loaded targets, and associating the overlapping degree with the corresponding first route which is left by the loaded target;

accumulating and calculating the overlapping degree associated with the first route from which the fed target leaves to obtain the sum of the overlapping degrees of the first route from which the fed target leaves;

taking the overlapping degree and the maximum overlapping degree of the first route from which each fed target leaves and the corresponding first route as a third route;

and manufacturing an avoidance route set based on each third route.

The working principle and the beneficial effects of the technical scheme are as follows:

when the evasion route set is manufactured, only one leaving route corresponding to each loaded target is needed to be ensured, overlapping between the first routes is allowed, the associated maximum overlapping degree and the corresponding first route left by the loaded target are selected as the third route, and the third route is brought into the evasion route set, so that the peripheral space of the robot is utilized to the maximum, the loading capacity is improved, and the system has higher applicability.

The invention provides a control system of an intelligent loading and unloading robot.A module for acquiring reminds a loading person to move a loading carrier to any target position block comprises:

determining the face position and the face orientation of the loading person based on the current first image;

constructing a first direction vector based on the face position and the face orientation;

acquiring equipment positions and equipment orientations of a plurality of first prompt equipment in a preset range around the robot;

constructing a second direction vector based on the device location and the device orientation;

calculating a first included angle between the first direction vector and the second direction vector;

selecting a first prompting device corresponding to the maximum first included angle as a second prompting device;

determining a third direction vector based on the device location of the second prompting device and the third center location of the target location block;

on the basis of a preset direction vector-guide information comparison library, comparing and determining first guide information corresponding to a third direction vector;

and controlling the second prompting device to display the first guide information.

The working principle and the beneficial effects of the technical scheme are as follows:

the mode that reminds material loading personnel to remove material loading carrier to arbitrary target location piece has one: remind the material loading personnel through first suggestion equipment, first suggestion equipment can be for display screen etc. for making things convenient for the material loading personnel to look over, sets up in higher department.

When the first prompting equipment is selected, the first direction vector and the second direction vector are introduced, the included angle between the first direction vector and the second direction vector is calculated, and the first prompting equipment corresponding to the maximum included angle is selected as the second prompting equipment. Typically, the angle is 180 ° when the loader is looking completely through the display. The device is oriented to be perpendicular to the display screen and faces outwards, a third direction vector and a direction vector-guiding information comparison library are introduced, and first guiding information corresponding to the third direction vector is determined in a comparison mode, for example: a 3D arrow pointing from the display screen location to the target location block. The direction vector-guide information comparison library is constructed by presetting guide information suitable for different direction vectors by a worker, and the direction of the display screen position to the target position block can be ensured to be suitable for ensuring that the guide information can indicate the direction. The accessibility and the vividness of the reminding are fully ensured. Vector construction and vector angle calculation belong to the category of the prior art and are not described in detail.

The invention provides a control system of an intelligent loading and unloading robot.A module for acquiring reminds a loading person to move a loading carrier to any target position block comprises:

acquiring projection ranges of a plurality of projection devices in a preset range around the robot;

when the projection range comprises the position area range of the target position block, acquiring a projection route corresponding to the projection equipment when the projection equipment projects preset second guide information in the position area range of the target position block;

determining whether an occlusion exists on the projection route based on the current first image;

if not, controlling any corresponding projection equipment to project second guide information in the position area range of the target position block.

The working principle and the beneficial effects of the technical scheme are as follows:

the mode that reminds the material loading personnel to remove the material loading carrier to any target position piece also has one: remind the material loading personnel through projection equipment, and the projection equipment can be a projection lamp and the like. However, the first prompting device and/or projecting device, etc. need to be chosen reasonably.

When the projection equipment is selected, the projection range is ensured to contain the position area range of the target position block, and the projection route is also ensured not to have a shielding object, so that the reminding accessibility is ensured. In addition, the preset second instruction information may be, for example: the frame is adapted to the position area range of the target position block, and the frame is internally provided with a key for moving the loading carrier to the frame! "is used herein.

The invention provides a control system of an intelligent loading and unloading robot.A control module controls the robot to transfer a workpiece in a loading area to an input end of a processing area or transfer the workpiece at an output end of the processing area to a preset unloading area, and when a manipulator of the robot is ready to clamp the workpiece, the best grabbing force of the workpiece is obtained, and the robot is controlled to grab the workpiece with the best grabbing force.

The working principle and the beneficial effects of the technical scheme are as follows:

generally, the grabbing force of the manipulator of the robot is fixedly set, so that the system can only be applied to feeding and discharging of one workpiece, and when the grabbing force is applied to other workpieces, the grabbing force needs to be set again. However, in an application scene, the types of general workpieces are more, the materials and the structures are different, the use is more complicated, and the applicability is low. Therefore, a solution is needed. This application is when the manipulator preparation clamp of robot gets the work piece, acquires the good dynamics of snatching of work piece, and control robot snatchs the work piece with the best dynamics of snatching, guarantees that the manipulator snatchs the work piece with suitable dynamics, promotes the suitability.

The invention provides a control system of an intelligent loading and unloading robot, wherein a control module acquires the optimal grabbing force of a workpiece, and the control system comprises:

acquiring a workpiece image of a workpiece;

determining structural information and materials of the workpiece based on the workpiece image;

extracting a plurality of structural features of the structural information based on a preset first feature extraction template;

training an optimal grabbing force determination model;

and inputting the material and the structural characteristics to the optimal grabbing force determination model based on the optimal grabbing force determination model, and determining the optimal grabbing force.

The working principle and the beneficial effects of the technical scheme are as follows:

when acquiring the workpiece image of the workpiece, the image acquisition device on the manipulator close to the workpiece can acquire the image. Based on the workpiece image, the structural information and material of the workpiece are determined, which can be realized based on image recognition technology. Introducing a preset first feature extraction template, extracting a plurality of structural features of the structural information, wherein the preset first feature extraction template can be: the structural features may be: how large and what structure, etc. An optimal grabbing force determining model is introduced, the optimal grabbing force determining model is based on the optimal grabbing force determining model, the optimal grabbing force is determined according to the material and the structural characteristics, and the determining efficiency and accuracy are improved.

The invention provides a control system of an intelligent loading and unloading robot.A control module trains an optimal grabbing force determination model, which comprises the following steps:

a plurality of grasping strength test records obtained from a big data platform and/or from the local;

preprocessing a gripping force test record;

taking the preprocessing result as a training sample, and carrying out model training on a preset neural network model;

after training is finished, taking the neural network model as an optimal grabbing force determination model;

wherein, control module carries out the preliminary treatment to snatching dynamics test record, includes:

acquiring record information of a grabbing force test record, wherein the record information comprises: source information and test procedure information;

extracting a plurality of recording features of the recording information based on a preset second feature extraction template;

acquiring a preset trigger feature library, and matching the recording features with first trigger features in the trigger feature library;

if the matching is in accordance with the first trigger characteristic, acquiring a trigger type of the first trigger characteristic in accordance with the matching, wherein the trigger type comprises: individual triggering and coordinated triggering;

when the trigger type is single trigger, acquiring a preset first verification scoring template corresponding to the matched first trigger characteristic;

when the trigger type is cooperative trigger, acquiring a preset cooperative trigger second trigger characteristic corresponding to the matched first trigger characteristic;

matching the recording features except the recording features matched and matched with the second trigger features in the recording features;

if the matching is in accordance with the preset verification marking template, acquiring a preset second verification marking template corresponding to the first triggering characteristic and the second triggering characteristic which are in accordance with the matching;

verifying and scoring the recording characteristics based on the first verification scoring template and the second verification scoring template, and accumulating scoring results to obtain a scoring sum;

if the score sum is less than or equal to the preset score sum threshold value, rejecting the corresponding grabbing force test record;

and after all the grabbing force test records needing to be removed are removed, integrating the remaining grabbing force test records to obtain a preprocessing result.

The working principle and the beneficial effects of the technical scheme are as follows:

"and/or" means that there are two parallel schemes for obtaining the record of the grabbing force test, and the schemes can be performed alternatively or together.

When the optimal grabbing force determination model is trained, firstly, a training sample needs to be obtained, the training sample can be a grabbing force test record, and the grabbing force test record is test data and test results of optimal grabbing force tests on workpieces of different materials and different structures. In order to improve the training quality of the optimal grabbing force determination model, the grabbing force test records are preprocessed, the preprocessed result is used as a training sample, and model training is carried out on a preset neural network model. The neural network model and the model training are both in the prior art and are not described in detail.

When the grabbing force test record is preprocessed, the grabbing force test record is verified from the dimension of the source and the dimension of the test process, and therefore the reliability of the source and the process quality of the test process are guaranteed.

And introducing a preset second feature extraction template, and extracting a plurality of recording features of the recording information, wherein the preset second feature extraction template can be recording features corresponding to different recording information, and the recording features can be: which source, what type of test, what test strategy to employ, etc.

The method is characterized in that a trigger feature library is introduced, a large number of first trigger features for representing what verification scoring templates are adopted to verify and score the recording features are arranged in the trigger feature library, and collection and library building can be performed in advance by a worker, so that the method can be realized. The trigger category of the first trigger feature is divided into individual trigger and cooperative trigger, and the individual trigger is that the first trigger feature can represent what verification scoring template is adopted to perform verification scoring on the record feature, for example: the method is characterized in that a website with a certain testing organization as a source is provided, a verification scoring template is provided with website credibility and corresponding scores, and collaborative triggering is realized by cooperatively representing what verification scoring template is adopted to verify and score recording characteristics by a first triggering characteristic and at least one other second triggering characteristic, for example: the characteristics are the experience value of the tester and the participation degree of the tester respectively, and the verification scoring template is a numerical value and a corresponding score obtained by endowing the participation degree as a weight to the experience value (multiplying the experience value and the experience value).

And respectively acquiring a corresponding first verification scoring template and a corresponding second verification scoring template based on different trigger types. The pertinence of the verification scoring and the verification scoring efficiency are improved to a great extent.

And verifying and scoring the recording characteristics based on the first verification scoring template and the second verification scoring template, and accumulating scoring results (scores of each verification scoring) to obtain a scoring sum. And if the score sum is less than or equal to the preset score sum threshold, removing the corresponding grabbing force test records, and integrating the rest grabbing force test records to obtain a preprocessing result, so that the training quality of the optimal grabbing force determination model is improved, and the applicability of the optimal grabbing force determination model for determining the optimal grabbing force according to the material and structural characteristics is improved.

The invention provides a control method of an intelligent loading and unloading robot, which comprises the following steps:

step 1: obtaining a feeding area;

step 2: controlling the robot to transfer the workpieces in the feeding area to the input end of the processing area, completing feeding, and after the processing is completed, controlling the robot to transfer the workpieces at the output end of the processing area to a preset discharging area;

step 1: obtain the material loading district, include:

continuously acquiring a first image in a preset range around the robot;

determining a first position block of a residual historical loading area in a preset range around the robot based on the first image;

determining the distribution of the feeding areas of the remaining historical feeding areas within a preset range around the robot based on the first position block;

acquiring the feeding state of a historical feeding area;

classifying the historical loading area into a loaded target and an unloaded target based on the loading state;

planning at least one first route of the loaded target to leave a preset range based on the first position block of the loaded target, the distribution of the loading area and a preset range map;

manufacturing an evasion route set based on the first route;

selecting a third position block which is idle and has no way for all second routes in the avoidance route set and has a first straight line distance between a first center position of the second position block and the robot smaller than or equal to a preset straight line distance from a plurality of second position blocks in the range map;

selecting a third position block corresponding to the minimum second straight-line distance in the second straight-line distances between the second center position of the first position block without the fed target and the first center position of the third position block as a target position block;

determining at least one loading person who enters a preset range around the robot and uses the loading carrier based on the current first image;

reminding a loading person to move the loading carrier to any target position block;

and determining that the loading personnel moves the loading carrier to the target position block based on a first image acquired after reminding the loading personnel to move the loading carrier to any target position block, and finishing acquisition as a loading area.

The invention provides a control method of an intelligent loading and unloading robot, which is used for manufacturing an evaded route set based on a first route and comprises the following steps:

acquiring the overlapping degree between a first route which is left by any one loaded target and a first route which is left by other loaded targets, and associating the overlapping degree with the corresponding first route which is left by the loaded target;

accumulating and calculating the overlapping degree associated with the first route from which the fed target leaves to obtain the overlapping degree sum of the first route from which the fed target leaves;

taking the overlapping degree and the maximum overlapping degree of the first route from which each fed target leaves and the corresponding first route as a third route;

and manufacturing an avoidance route set based on each third route.

It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.

Claims (9)

1. The utility model provides a control system of unloading robot in intelligence which characterized in that includes:

the acquisition module is used for acquiring the feeding area;

the control module is used for controlling the robot to transfer the workpieces in the feeding area to the input end of the processing area, the feeding is finished, and after the processing is finished, the robot is controlled to transfer the workpieces at the output end of the processing area to a preset blanking area;

the acquisition module acquires a loading area, including:

continuously acquiring a first image in a preset range around the robot;

determining a first position block of a residual historical loading area in the preset range around the robot based on the first image;

determining the distribution of the feeding areas of the remaining historical feeding areas in the preset range around the robot based on the first position block;

acquiring the feeding state of the historical feeding area;

classifying the historical loading area into a loaded target and an unloaded target based on the loading state;

planning at least one first route for the loaded target to leave the preset range based on a preset range map, the first position block of the loaded target and the loading area distribution;

making an avoidance route set based on the first route;

selecting a third position block which is idle and has no way for all second routes in the avoidance route set and has a first straight line distance between a first center position of the second position block and the robot smaller than or equal to a preset straight line distance from a plurality of second position blocks in the range map;

selecting the third position block corresponding to the minimum second straight-line distance in the second straight-line distances between the second center position of the first position block of the unloaded target and the first center position of the third position block as a target position block;

determining at least one loading person who enters the preset range around the robot and uses the loading carrier based on the current first image;

reminding the loading personnel to move the loading carrier to any one target position block;

and determining that the loading personnel moves the loading carrier to any target position block based on the first image acquired after reminding the loading personnel to move the loading carrier to the target position block, and using the first image as a loading area to finish acquisition.

2. The control system of claim 1, wherein the obtaining module is configured to make an evasive route set based on the first route, and comprises:

acquiring the overlapping degree between a first route which is left by any one loaded target and a first route which is left by other loaded targets, and associating the overlapping degree with the corresponding first route which is left by the loaded target;

accumulating and calculating the overlapping degree associated with the first route which the fed target leaves to obtain the overlapping degree sum of the first route which the fed target leaves;

taking the overlapping degree and the maximum overlapping degree of the first route which each loaded target leaves and the corresponding first route as a third route;

and manufacturing an avoidance route set based on each third route.

3. The system as claimed in claim 1, wherein the module for reminding the loader to move the loader to any one of the target position blocks comprises:

determining the face position and the face orientation of the loading person based on the current first image;

constructing a first direction vector based on the face position and the face orientation;

acquiring the equipment positions and equipment orientations of a plurality of first prompt equipment in the preset range around the robot;

constructing a second direction vector based on the device location and the device orientation;

calculating a first included angle between the first direction vector and the second direction vector;

selecting the first prompting device corresponding to the largest first included angle as a second prompting device;

determining a third directional vector based on the device location of the second prompting device and a third center location of the target location block;

on the basis of a preset direction vector-guide information comparison library, comparing and determining first guide information corresponding to the third direction vector;

and controlling the second prompting equipment to display the first guide information.

4. The system as claimed in claim 1, wherein the module for reminding the loader to move the loader to any one of the target position blocks comprises:

acquiring the projection ranges of a plurality of projection devices in the preset range around the robot;

when the projection range contains the position area range of the target position block, acquiring a projection route corresponding to the projection equipment when the projection equipment projects preset second guide information in the position area range of the target position block;

determining whether an obstruction exists on the projection route based on the current first image;

if not, controlling any corresponding projection equipment to project the second guide information to the position area range of the target position block.

5. The system as claimed in claim 1, wherein the control module controls the robot to transfer the workpiece in the loading area to the input end of the processing area or transfer the workpiece at the output end of the processing area to a preset unloading area, and when the robot arm is ready to grip the workpiece, the control module obtains an optimal gripping force for the workpiece and controls the robot to grip the workpiece with the optimal gripping force.

6. The control system of claim 1, wherein the control module obtains the optimal gripping force of the workpiece, and comprises:

acquiring a workpiece image of the workpiece;

determining structural information and material of the workpiece based on the workpiece image;

extracting a plurality of structural features of the structural information based on a preset first feature extraction template;

training an optimal grabbing force determination model;

and inputting the material and the structural characteristics to the optimal grabbing force determination model based on the optimal grabbing force determination model, and determining the optimal grabbing force.

7. The control system of claim 6, wherein the control module trains an optimal grasping force determination model, and comprises:

a plurality of grasping strength test records obtained from a big data platform and/or from the local;

preprocessing the grabbing force test record;

taking the preprocessing result as a training sample, and carrying out model training on a preset neural network model;

after training is finished, the neural network model is used as an optimal grabbing force determination model;

wherein, control module is to snatch dynamics test record carries out the preliminary treatment, include:

acquiring record information of the grabbing force test record, wherein the record information comprises: source information and test procedure information;

extracting a plurality of recording features of the recording information based on a preset second feature extraction template;

acquiring a preset trigger feature library, and matching the recording feature with a first trigger feature in the trigger feature library;

if the matching is in accordance with the first trigger characteristic, acquiring a trigger type of the first trigger characteristic in accordance with the matching, wherein the trigger type comprises: individual triggering and coordinated triggering;

when the trigger type is single trigger, acquiring a preset first verification scoring template corresponding to the matched first trigger characteristic;

when the trigger type is cooperative trigger, acquiring a preset cooperative trigger second trigger characteristic corresponding to the matched first trigger characteristic;

matching the recording features except the recording features matched and matched with the second triggering features in the recording features;

if the matching is in accordance with the preset verification scoring template, acquiring a preset second verification scoring template corresponding to the first triggering characteristic and the second triggering characteristic which are in accordance with the matching;

verifying and scoring the recording characteristics based on the first verification scoring template and the second verification scoring template, and accumulating scoring results to obtain a scoring sum;

if the score sum is less than or equal to a preset score sum threshold value, rejecting the corresponding grabbing force test record;

and after all the grabbing force test records needing to be removed are removed, integrating the remaining grabbing force test records to obtain a preprocessing result.

8. A control method of an intelligent loading and unloading robot is characterized by comprising the following steps:

step 1: obtaining a feeding area;

step 2: controlling the robot to transfer the workpieces in the feeding area to the input end of the processing area, completing feeding, and after the processing is completed, controlling the robot to transfer the workpieces at the output end of the processing area to a preset discharging area;

the step 1: obtain the material loading district, include:

continuously acquiring a first image in a preset range around the robot;

determining a first position block of a residual historical loading area in the preset range around the robot based on the first image;

determining the distribution of the feeding areas of the remaining historical feeding areas in the preset range around the robot based on the first position block;

acquiring the feeding state of the historical feeding area;

classifying the historical loading area into a loaded target and an unloaded target based on the loading state;

planning at least one first route for the loaded target to leave the preset range based on a preset range map, the first position block of the loaded target and the loading area distribution;

making an avoidance route set based on the first route;

selecting a third position block which is idle and has no way for all second routes in the avoidance route set and has a first straight line distance between a first center position of the second position block and the robot smaller than or equal to a preset straight line distance from a plurality of second position blocks in the range map;

selecting the third position block corresponding to the minimum second straight-line distance in the second straight-line distances between the second center position of the first position block of the unloaded target and the first center position of the third position block as a target position block;

determining at least one loading person who enters the preset range around the robot and uses the loading carrier based on the current first image;

reminding the loading personnel to move the loading carrier to any one target position block;

and determining that the loading personnel moves the loading carrier to any target position block based on the first image acquired after reminding the loading personnel to move the loading carrier to the target position block, and using the first image as a loading area to finish acquisition.

9. The method for controlling the intelligent loading and unloading robot as claimed in claim 8, wherein the step of making the evasive route set based on the first route comprises:

acquiring the overlapping degree between a first route which is left by any one loaded target and a first route which is left by other loaded targets, and associating the overlapping degree with the corresponding first route which is left by the loaded target;

accumulating and calculating the overlapping degree associated with the first route which the fed target leaves to obtain the overlapping degree sum of the first route which the fed target leaves;

taking the overlapping degree and the maximum overlapping degree of the first route which each loaded target leaves and the corresponding first route as a third route;

and manufacturing an avoidance route set based on each third route.

Images